How much random variation in motor inrush peak current due to closing

[electricpete]

We know that motor peak inrush current (first cycle) can vary randomly based on the phase angle of the voltage at the time of closiong.

Let's say I start the motor many times and each time record the single highest magintude peak (either polarity, any phase) for each start.

Assuming ideal conditions (simultaneous contact closure on all 3 phases without bounce, balanced voltage, no motor assymetries), what would you estimate the variability of these recorded numbers as as a result of random variation of voltage phase at time of closing?

From a simple R/L circuit model, I estimated around 6% variability (max peak minus min over average) at pf = 0.15. Spreadsheet attached (don't look at the cell formula's... they are uglier than they need to be). Seems like surprisingly low variability. Maybe I made an error?

What do you estimate?


=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[dpc]

But aren't you really mainly concerned about the magnitude of the peak (maximum case) versus the minimum (no transient inrush) ? The average value is really of little interest in terms of setting the instantaneous trip.

 

[electricpete]

I appreciate that you are looking to understand or solve whatever is the overall question/problem. I have lots of questions/problems, but want to attack them in well-defined individual chunks. (Many more chunks will follow, I guarantee).

This particular chunk - we recorded a multitude of starts. I was presented summary data consisting of a single number for each start - the highest peak on any of the phases during that start. There was considerable variability in those particular numbers. I did an analysis (attached) to see what is expected variability under these particular assumptions (balance etc). The results suprpised me (low variability). I figure they might suprise others as well. I want to make sure that particular analysis is correct. If so, it helps us BEGIN to answer some questions I have been asked like: How many starts do I need to do to capture the worst case? If I record 3 starts and want to set my breaker comfortably above the behavior observed during those starts, how much additional margin should I include to allow for closing time variation (yes, there is a ton of other margin to be accounted for, including the balance assumption which I think will be a big factor in explaining the variability... but it is not the question here).

So the question is unchanged, do you agree with my analysis/conclusion as described in my first post, under the assumptions given ?

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[waross]

Hi Pete;
When you start a motor one phase is always in a position to give a large inrush.
I couldn't load your spreadsheet but I agree that there will be a relatively small variation between the maximum currents of the highest phase.
The ripple of a three phase bridge rectifier is much less than the ripple of a single phase bridge rectifier. I believe there may be a similarity.

Bill
--------------------
"Why not the best?"
Jimmy Carter